PROMOTING TAX COMPLIANCE IN KOSOVO WITH BEHAVIORAL … · 2019. 3. 5. · 03 Kosovo Promoting Tax...

PROMOTING TAX COMPLIANCE IN KOSOVO WITH BEHAVIORAL INSIGHTS

Marco Hernandez - Jonathan Karver - Mario Negre - Julie Perng

MARCH 2019P

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

ed

02

Kosovo: Promoting Tax Compliance with Behavioral Insights

© 2019 International Development Agency / The World Bank 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org

This work is a product of the staff of The World Bank in cooperation with the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH and with external contributions. The findings, in-terpretations, and conclusions expressed in this work do not necessarily reflect the views of the GIZ, The World Bank, its Board of Executive Directors, or the governments they represent.

The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judg-ment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries.

Rights and Permissions

The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: [email protected].

Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this trans-lation.

Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank.

Third-party content—The World Bank does not necessarily own each component of the content con-tained within the work. The World Bank therefore does not warrant that the use of any third-party–owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is need-ed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images.

03


Acknowledgments

This report has been prepared at the request of the Government of Kosovo. The team is exceedingly grateful to the Ministry of Finance of Kosovo and the Tax Administration of Kosovo (TAK) for their fruitful cooperation in this initiative and for providing knowledge, data, and advice.

This report was prepared jointly by staff at the World Bank and the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) and benefited from contributions from staff at the Tax Administration of Kosovo. The initiative was financed by Deutsche Gesellschaft für Interna-tional Zusammenarbeit (GIZ) GmbH, on behalf of the German Federal Ministry for Econom-ic Cooperation and Development (BMZ) and was carried out in a partnership between the World Bank’s Mind, Behavior, and Development (eMBeD) unit and the GIZ Public Finance Project in Kosovo.

The World Bank team was led by Jonathan Karver (Research Analyst), Marco Antonio Her-nandez Ore (Program Leader and Lead Economist), Mario Negre (Consultant), and Julie Perng (Research Analyst). The extended team included Abigail Dalton (Operations Officer), Monica Robayo (Economist), Asli Senkal (Economist), Agim Demukaj (Economist), and Lundrim Aliu (Communications Officer). The World Bank team worked under the overall guidance of Linda Van Gelder (Regional Director for the Western Balkans), Marco Mantovanelli (Country Man-ager for Kosovo and the former Yugoslav Republic of Macedonia), Renos Vakis (Co-Head of the Mind, Behavior, and Development eMBeD unit), Oscar Calvo-Gonzalez (Practice Manager and Supervisor of the Mind, Behavior, and Development eMBeD unit), Carlos Silva-Jauregui (Acting Practice Manager, Poverty and Equity), and Gallina Vincelette (Practice Manager, Macroeco-nomics, Trade, and Investment).

The GIZ team was led by Kadri Jõgiste (Project Leader, Public Finance Project), and included Njomza Ilazi (Advisor, Public Finance Project), Ramadan Islami (Project Coor-dinator, Public Finance Project), Sylvia Van Ziegert (Senior Advisor, Sectoral Depart-ment) and Stefanie Rauscher (Senior Advisor, Sectoral Department). The GIZ team worked under the overall guidance of David Oberhuber (Country Director, Kosovo and the former Yugoslav Republic of Macedonia) and Henning Twesten (Principal Economist, Europe, Mediterranean Region, Central Asia Department).

04


We would especially like to thank the dedicated staff at the Tax Administration of Kosovo for their continuous support and contributions throughout the duration of the project. We are especially grateful to Ilir Murtezaj (Director General), Sakip Imeri (Former Director General), Fjolla Muja (Directorate for International Cooperation), Ardita Haxhnikaj Demi (Taxpayer Education and Service Department), Valdet Shala (Directorate for Debts and Tax Account Management), Njomëza Alaj (Information and Communication Technology De-partment), Kimete Bilalli (Information and Communication Technology Department), and Adelina Sutaj (Call Center).

Excellent organizational and administrative assistance was provided by Ivana Bojic. Robert Zimmermann provided assistance in editing, and Lorena Guedes in designing the report.

05


ContentsAcknowledgments ....................................................................................................................03

Abbreviations and Acronyms .................................................................................................05

Executive Summary ..................................................................................................................06

Introduction................................................................................................................................10

Tax Compliance in Kosovo ......................................................................................................15

Behavioral Interventions in Tax Compliance: Three Experiments ..................................21

Summary of Results ..................................................................................................................25

4.1 Letter Reminders to Taxpayers Late in

Submitting their PIT Declarations ............................................................26

4.2 E-mail Reminders to Firms Subject to the Value Added Tax ..................28

4.3 SMS Reminders to Firms Subject to the Value Added Tax .....................30

Process Makes Perfect: Lessons Learned from Implementation ....................................33

5.1 Implementation Successes .............................................................................34

5.2 Implementation Challenges ...........................................................................36

5.3 Opportunities Ahead .......................................................................................38

Conclusion ..................................................................................................................................42

References ..................................................................................................................43

Notes............................................................................................................................45

Annexes .......................................................................................................................47

Abbreviations and Acronyms

EU European UnionGDP Gross Domestic ProductGIZ Gesellschaft für Internationale ZusammenarbeitPIT Personal Income TaxRCT Randomized Controlled TrialSMS Short Messaging ServiceTAK Tax Administration of KosovoVAT Value Added Tax

06


Executive Summary

As in many countries, tax collection is a develop-ment challenge in Kosovo. Kosovo is one of the poorest and youngest countries in Europe in terms of gross domestic product (GDP) per capita and both demographics and statehood. The lack of an independent monetary policy—given that Kosovo has adopted the euro as the national currency—means that ensuring the sustainability of fiscal policy is critical. However, limited tax revenues hamper the government’s ability to address eco-nomic cycles. Between 2011 and 2017, total gov-ernment revenue amounted to about 14 percent of GDP, below the average of 19 percent among countries in Europe and Central Asia. Unlike other countries that collect significant nontax revenues (for example, from natural resources), the Koso-var government relies on taxes for more than 85 percent of its revenues. Mobilizing tax revenues is therefore critical from both an efficiency perspec-tive (to ensure that taxes are collected at the low-est cost for the public administration) and an equi-ty perspective (to ensure that all taxpayers comply by declaring and paying their due taxes).

The Tax Administration of Kosovo (TAK) re-quested assistance from the World Bank and the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) to address this challenge using an evidence-based approach. To this end,

the World Bank and GIZ applied behavioral in-sights to promote tax compliance among specif-ic groups of taxpayers. Three experiments were designed, implemented, and evaluated in 2018 that involved sending behaviorally informed re-minders using letters, e-mails, and short mes-saging service (SMS) messages to various groups of taxpayers to induce timely and honest decla-rations and payments. The short-term objective of these trials was to increase the number and timeliness of tax declarations.

Simple, behaviorally designed messages were effective in inducing tax declaration in Kosovo. Messages helped raise the tax declaration rate by an average of around 3 percentage points during a period of between four and six weeks. Among personal income tax (PIT) declarations, this represents a 59 percent increase in compli-ance, equivalent to over 200 more annual tax declarations among participants. The likelihood of payment rose in many instances, and no sig-nificant difference was found in the amounts of taxes paid.

Lessons from the tax experiments in Koso-vo highlight the benefits of rigorous impact evaluation and the need to establish process-es that help integrate tax collection functions

07


and data systems. The effectiveness of tax col-lection efforts is a function of a myriad of fac-tors, many of which are context specific. Estab-lishing a process to test tax collection efforts, learn from them, and adapt systems based on the lessons is at least as important as the impact of the interventions on revenue mobilization, which represents a longer-term goal that itself is a function of changes in declaration behav-ior. The three experiments conducted in Kosovo generated lessons in two main areas: the quality of information systems and the existing com-munication infrastructure. Notwithstanding the resources available to the tax administration to reach out to taxpayers and to monitor tax com-pliance, an inability to locate and contact tax-payers can make or break a tax collection initia-tive. In other words, a communication campaign to remind taxpayers to declare on time and pay their due taxes will only be effective if taxpay-ers receive and read the reminders. Hence, the implementation of these tax interventions in Kosovo provides a critical learning experience and capacity-building exercise for policy makers.

It is recommended that similar tax collection efforts in Kosovo in the future investigate ad-ditional communication channels and popula-tions of interest. The former might involve web-

site-based messaging or phone calls through a call center. The populations of interest might involve unregistered income earners and busi-nesses. Future tax experiments might also learn from the challenges outlined in this report by in-vesting in the design and in the process equally and by relying closely on the results from each trial to test, learn, adapt, and retest.

08


Përmbledhje ekzekutive

Si në shumë vende, mbledhja e tatimit është sfidë zhvillimore në Kosovë. Kosova është një nga vendet më të varfra dhe më të rejat (në as-pektin e demografisë dhe të shtetësisë) në Evropë. Mungesa e një politike monetare të pavarur - duke qenë se Kosova e ka aplikuar Euron si monedhën e saj - do të thotë se sigurimi i qëndrueshmërisë së politikës fiskale është kritik. Megjithatë, të hyrat e kufizuara nga tatimi pengojnë aftësinë e qeverisë për t’i adresuar ciklet ekonomike. Midis viteve 2011 dhe 2017 të hyrat e teresishme të qeverisë në Kosovë arritën në rreth 14 për qind të Bruto Prodhimit Vendor (BPV), nën mesataren prej 19 për qind për vendet e Evropës dhe të Azisë Qendrore. Ndryshe nga vendet tjera që mbledhin sasi të konsiderueshme nga të hyrat jo-tatimore (p.sh., nga burimet natyrore), qeveria e Kosovës mbështetet në të hyrat tatimore për më shumë se 85 për qind të të hyrave të saj. Në këtë kontekst, mobilizimi i të hyrave tatimore në Kosovë është i një rëndësie të veqantë: si nga perspektiva e efi-kasitetit (për të siguruar se tatimet mblidhen me koston më të ulët për administratën publike) ashtu edhe nga perspektiva e ekuitetit (për të siguruar se të gjithë tatimpaguesit janë në pajtim me deklar-imin dhe pagesën e tatimeve).

Administrata Tatimore e Kosovës (ATK) kërkoi asistencë nga Banka Botërore dhe Deutsche Ge-

sellschaft fuer Internationale Zusammenarbeit (GIZ) për ta adresuar këtë sfidë duke e përdorur një qasje të bazuar në dëshmi. Për këtë qëllim, Banka Botërore dhe GIZ aplikuan njohuritë rreth sjelljes së njerëzve për ta promovuar pagesën e tatimeve në mesin e grupeve te veqanta të tatim-paguesve. Tri eksperimente u dizajnuan, u zbat-uan dhe u vlerësuan në vitin 2018, që përfshinin dërgimin e mesazheve përkujtuese bazuara ne informatat rreth sjelljes (përmes letrave, postet elektronike dhe mesazheve te shkurtera SMS) grupeve të ndryshme të tatimpaguesve për ta nxitur deklarimin e saktë dhe pagesën me kohë. Qellimi afatshkurter i ketyre tentativave ishte rritja e numrit dhe permiresimi i afatit kohor ne deklarimet e tatimit.

Mesazhet e thjeshta, të disenjuara sipas sjelljeve ishin efikase në rritjen e deklarimeve tatimore në Kosovë. Mesatarisht, mesazhet ndihmuan në rritjen e normës së deklarimit të tatimit në rreth 3 pikë përqindjeje në deklarimin e tatimeve në të ardhura përsonale (TAP), kjo paraqet 59 përqind rrritje e përmbushjes, e barabartë me 200 deklarime tatimore më shumë mes pjesmarrësve - gjatë periudhës prej rreth 4-6 javëve. Gjasat e pagesës ishin dukshëm më të larta në shumicën e rasteve, dhe nuk gjetëm dallime statistikisht të rëndësishme në lidhje me shumën e tatimeve

09


të paguara.

Mësimet nga eksperimentet tatimore në Kosovë nxjerrin në pah përfitimet nga vlerësimi rigoroz i ndikimit dhe nevojën për të vendosur procese që ndihmojnë në integrimin më të mirë të funk-sioneve të mbledhjes së tatimeve dhe të siste-meve të të dhënave. Efektiviteti i përpjekjeve në grumbullimin e tatimeve është një funksion i disa faktorëve, shumë prej së cilëve janë specifikë në kontekst. Krijimi i një procesi për t’i testuar përp-jekjet e mbledhjes së tatimeve, për të mësuar prej tyre, dhe për t’i përshtatur sistemet në bazë të këtyre mësimeve është po aq e rëndësishme, sa edhe ndikimi që këto ndërhyrje mund të kenë në mobilizimin e të hyrave, e që paraqet një qëllim më aftgjatë. Nga tre eksperimentet e zhvilluara në Kosovë, janë identifikuar një numër i mësimeve të nxjerra në dy fusha kryesore: cilësia e sistemeve të informacionit dhe infrastruktura ekzistuese e komunikimit. Pavarësisht burimeve në dispozi-cion të administratës tatimore pë të arritur tek tatimpaguesit dhe për të monitoruar përmbush-jen e tatimeve, pamundësia për t’i kontaktuar ose gjetur tatimpaguesit mund ta bëjë ose ta prishë nismën për mbledhjen e tatimeve. Thënë thjesht, një fushatë komunikimi ku autoriteti tatimor i përkujton tatimpaguesit që t’i paguajnë tatimet e tyre do të jetë efektiv vetëm nëse tatimpaguesit

i marrin dhe i lexojnë këto porosi përkujtuese. Prandaj, implementimi i ndërhyrjeve tatimore në Kosovë siguron një përvojë kritike mësimore dhe për ndërtim të kapaciteteve për politikëbërësit.

Rekomandohet që përpjekjet e ngjashme për mbledhjen e tatimeve në Kosovë në të ardhmen të hulumtojnë për kanale shtesë të komunikimit si dhe grupeve të ndryshme të interesit. Të para mund të marrin parasysh porositë e dërguara përmes faqeve të internetit, ose telefonatat e bëra nga një qendër e telefonatave. Të fundit mund të përfshijnë fituesit e të ardhurave dhe bizneset e paragjistruara. Eksperimentet e ard-hshme lidhur me tatimet mund të mësojnë nga sfidat e theksuara në këtë raport duke investuar barabart si në proces edhe në dizajn dhe të ba-zohen në rezultatet nga çdo provë për të testuar, mësuar, përshtatur dhe ri-testuar.

10


1.Introduction

Kosovo is one of the poorest and youngest countries in Europe in terms of gross do-mestic product (GDP) per capita and both demographics and statehood.26 The youth of the country represent a prospect for future prosperity, but are hampered by structural is-sues, such as informality. Coupled with the population’s perceptions of institutional cor-ruption, years of mistrust toward authorities preceding the formation of the republic, and low satisfaction with public service deliv-ery, collecting taxes represents a substantial challenge (World Bank 2017).27 Mobilizing tax revenue efficiently and fairly is a priority of the government of Kosovo. Tax mobiliza-tion is crucial to ensuring sustainable devel-opment and fostering independence from foreign aid over the long term (Mascagni 2015). Indeed, the 2015 Addis Ababa Action Agenda, Financing for Development, calls on countries to step up their efforts to mobilize domestic resources, recognizing that much of the public financing necessary to achieve the Sustainable Development Goals must be generated domestically.1 In Kosovo, taxes are the most important source of public rev-

enues, accounting for about 88 percent of total revenues and grants during 2014 and 2018. They are thus critical in financing public goods and services, such as education, health care, and infrastructure. Moreover, given that Kosovo uses the euro as a unit of mone-tary exchange, the government also has only limited autonomy in monetary policy. Fis-cal policy—the use of government revenues and expenditures—is thus its main tool for influencing the economy. The efficient mobi-lization of tax revenues is therefore essential to ensuring economic sustainability.

Tax collection in Kosovo has expanded in recent years, but a number of challeng-es remain, and there is ample room for improvement. Tax revenues as a share of GDP rose from 21.1 percent in 2008, when the country became independent, to an es-timated 23.7 percent in 2017. Despite this progress, there is ample room to improve tax mobilization in Kosovo, given that tax reve-nues as a share of GDP are lower than in peer economies in Europe and in some economies at similar levels of income (figure 2, panel

11


b). For example, while the average tax revenue as a share of GDP between 2014 and 2017 in Kosovo stood at 22.4 percent, it stood at 24.4, 26.1, and 35.4 percent in Albania, Montenegro, and Serbia, respectively. The challenge of low tax compliance is especially acute in the case of the value added tax (VAT) and the personal income tax (PIT) because of a combination of tax eva-sion and weak enforcement capacity, that is, the inability to pursue noncompliant taxpayers. For instance, a World Bank (2014) study estimated the size of the large tax gaps in the VAT, PIT, and the corporate income tax, wherein the VAT gap was estimated at about 34 percent of collections; the PIT gap was about three times the amount collected; and the corporate income tax gap was estimated at about 17 percent of collections. Low levels of tax compliance have negative con-sequences. For the private sector, for example, tax evasion reduces collections and also creates an uneven playing field that concedes excessive space to politically connected firms. This nega-tively affects competition and productivity. Ul-timately, perceptions that tax compliance is low, whether they are correct or not, can influence re-ality by changing social norms around taxes. This is because taxpayers who believe others are not tax compliant may choose to avoid paying taxes themselves (Torgler 2007).

Traditional measures to boost government revenues include changes in tax legislation and reforms in tax administration; however, such measures can be costly, are often politically difficult to negotiate, and can take time to im-plement. Encouraging individuals to declare and

pay their taxes on time has involved a combina-tion of incentives (carrots) and sanctions (sticks). The latter have traditionally been the focus of tax collection strategies, that is, tax enforcement pro-cedures, including fees and penalties. Traditional means of enforcing compliance generally involve individual audits and residential or business visits, both of which require a level of human resourc-es that is difficult to finance and may even involve changes in policy, which are commonly infeasible. In Kosovo, moreover, limited budgetary resources and limited institutional capacity for reform in tax administration make investment difficult in tradi-tional enforcement mechanisms such as audits. It is therefore important also to consider carrots, for example, reminders, in the design and imple-mentation of tax collection strategies because these may be effective and less costly. How might one improve tax collection in Kosovo at low cost and without changing legislation? Evidence-based strategies that respect the principles of testing, learning, and adapting through the use of rigorous analysis may be the answer.

Behaviorally informed interventions represent an innovative, evidence-based approach to improving tax compliance and the efficiency of tax administration. They embody the social, psychological, and economic factors that affect how people think and act. They aim to identify efficiency gains given an existing bureaucratic and technological environment. Behaviorally in-formed policy can provide creative solutions to difficult challenges, often at low cost. This is true of tax collection, wherein compliance among in-dividuals and firms is rarely only a function of the

12


expected probability of receiving an audit and typically depends also on social norms and the framing of obligations, in addition to an evalua-tion of risk. The World Bank has also been ex-ploring the application of behavioral insights in various policy spaces. World Development Report 2015: Mind, Society, and Behavior (World Bank 2015) highlights that the application of behavioral science has the potential to amplify the effective-ness of public policy. People think automatically and socially and often use mental models that are unconscious. In other words, they apply heuris-tics and shortcuts that are particular to each con-text. They also tend to think in terms of stories or narratives rather than data points. These insights can help policy makers more closely align their communication strategies with the behavior of citizens. Policy makers have historically struggled to approve and implement tax reforms because of political opposition and bureaucratic inertia. Nonregulatory interventions rooted in the exper-imental methodology of behavioral economics can provide a valuable alternative to more con-ventional reform programs.

Examples from international experience un-derline the value of the behavioral approach to promote tax compliance, and the importance of establishing a process for testing, learning, and adapting.2 Traditional approaches to tax compli-ance, which revolve around deterrence messages like informing taxpayers about the probability of an audit, for example, represent a sizable portion of this literature (see, for example, Carrillo, Pomer-anz, and Singhal 2017). However, recent stud-ies emphasizing how psychological factors take

a major part in taxpayer behavior demonstrate that communication is a fundamental tool for im-proving compliance. By highlighting the benefits of taxation; eliciting notions of fairness, respon-sibility, and morality; emphasizing social norms; and framing information differently, taxpayer be-havior can be nudged in a way that uncovers bot-tlenecks unaddressed by traditional notions of deterrence (see Hallsworth et al. 2017; Hernan-dez et al. 2017; Brockmeyer et al., forthcoming). Evidence from a wide range of countries demon-strates the advantage of behavioral interventions to promote tax compliance because these gener-ally involve modifications to existing systems and processes and thus tend to be cost-effective and politically feasible. In recent years, the impact of behavioral interventions has been measured us-ing randomized controlled trials (RCTs), which are now widely applied in medicine, business, and in-ternational development. For example, the effec-tiveness of tax reminders has been tested using RCTs in a number of countries, including Argenti-na, Australia, Austria, Chile, Costa Rica, Denmark, Germany, Guatemala, Israel, Peru, Poland, Swit-zerland, the United Kingdom, the United States, and Venezuela.3 Common to these RCTs is their reliance on existing data collected and managed by tax authorities, leading to rapid, low-cost im-plementation.

Behavioral interventions are easily replicable and scalable, stimulating a process of adaptive learning. The success of behavioral messages meant to incentivize greater tax compliance has been clearly documented in recent research. Pi-oneering research by the Behavioural Insights

13


Team (BIT) and the tax authority in the United Kingdom highlights the efficacy of behaviorally informed messages communicated by standard mail. Messages invoking social norms, such as the average number of people who pay taxes on time, boosted the payment of declared tax liabilities by up to 5.1 percentage points within 23 days of delivery relative to a control group (Hallsworth et al. 2017). The World Bank and the BIT also worked closely with tax authori-ties in Guatemala to design better communica-tion strategies using these and other behavioral insights (Kettle et al. 2016). They found that a letter relying on a deliberate choice (omission versus commission) helped raise the average amount paid per taxpayer by $17.95 (269 per-cent). Undoubtedly, the effectiveness of these behavioral messages is not universal, and con-text is key. Merely because an intervention is successful in one setting does not guarantee success in another. Thus, while evidence on oth-er countries can provide insights to inform the design of behavioral interventions, there is still a need to experiment to determine whether be-havioral insights can promote tax compliance in Kosovo.

The Tax Administration of Kosovo (TAK) has begun to apply behavioral insights to promote tax compliance. In 2018, through a partnership between the World Bank and the Gesellschaft für Internationale Zusammenarbeit (GIZ), TAK carried out the first three RCTs to improve tax compliance, that is, to increase the number and timeliness of tax declarations. These tar-geted taxpayers subject to PIT and VAT across

the country and tested the impact of reminder messages on tax compliance outcomes. Fig-ure 1 displays a timeline of activities conduct-ed during the collaboration in 2018. A behav-ioral diagnostic was carried out in February 2018, and all three trials were designed and implemented between March and September. The diagnostic was meant to inform the design of the trials, specifically, the method, content, and tone of the messages sent to taxpayers. This was accomplished through a combination of semi-structured and unstructured interviews with taxpayers and various stakeholders, re-spectively, as well as conversations with units within the tax administration. This formative research keyed on the success of the model that views the taxpayer as client and that emphasiz-es voluntary tax compliance through a positive tone. This ultimately informed the decision to promote positively framed messages, such as the benefits of paying taxes, rather than negatively framed messages, such as deterrence.

This policy note summarizes the findings of these tax collection activities in Kosovo and fo-cuses on two key outcomes: the effectiveness of the trials in terms of the level of declarations (the impacts) and the process for carrying out the trials (the implementation). Experimental trials have generally concentrated on measur-able results, that is, how much additional rev-enue can be attributed to some policy or how the declaration rate may be increased by tweak-ing an existing framework. However, in Koso-vo, the focus was on a dimension that gener-ally receives less attention in the literature: the

14


effectiveness of implementation as an exercise in assessing the operations of a tax adminis-tration, building capacity, and learning lessons that may be extracted for future trials. Thus, a key outcome of the trials has been the discov-ery that nearly two-thirds of the entries in the

taxpayer registry are associated with incorrect addresses or no addresses at all. The limited in-formation on taxpayer addresses is especially rel-evant because it clearly affects the analysis of the reaction of the population of interest to the inter-ventions and in changing the mindsets of policy

F E B2 0 1 8 D i a g n o s t i c o f t a x c o m p l i a n c e

M A R2 0 1 8

I m p l e m e n t a t i o n o f P I T t r i a l

M AY2 0 1 8

B ra i n s t o r m i n g f o r VAT t r i a l s , e a r l y r e s u l t s P I T t r i a l

J U L2 0 1 8 I m p l e m e n t a t i o n o f VAT e - m a i l t r i a l

S E P2 0 1 8

I m p l e m e n t a t i o n o f VAT S M S t r i a l

D E C2 0 1 8 R e s u l t s d i s s e m i n a t i o n

TestLearnAdapt

Re-test !

Figure 1. TIMELINE OF ACTIVITIES

15


makers and implementers with regard to their tax clients.

2. Tax Compliance in Kosovo

Kosovo has a simple tax system and relatively low tax rates. A World Bank (2014) report ti-tled “Kosovo Public Revenues: Tax Policies, Tax Evasion, and Tax Gaps” presents a detailed as-sessment of the Kosovar tax system. The coun-try’s VAT rate is flat, at 18 percent, and the PIT rate is progressive, at 0, 4, 8, and 10 percent on income tax brackets of €0–€80, €81–€250, €251–€450, and over €450 a month.4 Kosovar tax legislation is broadly aligned with European Union (EU) standards. Tax rates are low com-

pared with other countries in Europe and Cen-tral Asia. Moreover, a feature of the World Bank (2014) report is the analysis of the finding that the tax system is highly dependent on customs taxes collected at the border and that there is a need to shift from border taxes to domestic sources of revenue, including the VAT, PIT, the corporate income tax, and property taxes.

Tax compliance has generally improved over time, but the collection of VAT and PIT reve-nues remains problematic. Average tax revenues in Kosovo over the past seven years represent 14 percent of total revenues. Although the tax-to-GDP ratio is similar in Kosovo and other coun-tries in the Western Balkans, tax revenue is over-whelmingly sourced from indirect taxation (more than 80 percent), particularly VAT, excise, and in-ternational trade taxes. Meanwhile, PITs and tax-es on profit account for 14 percent of tax revenue (less than 4 percent of GDP), about half the share in other countries in the Western Balkans and in the EU accession countries.

16


The heavy dependence on indirect taxation is consistent with the import intensity and transfer dependence characteristic of the Kosovar econo-my. Indeed, remittances reach 14 percent of GDP, higher than in all other Western Balkan countries. The share of the imports of goods and services in GDP was around 52.5 percent in 2017, while the trade deficit was 25.9 percent of GDP (World Bank 2017). The government budget deficit of 6.3 percent in 2017 and the mounting public debt, at 16.6 percent of GDP in the same year, both underscore the need to raise domestic revenues to engage along a financially sustainable path.5 The low revenue from income taxes derives from a combination of factors, mainly relatively low tax rates, the narrow tax base (because of low labor force participation and high unemployment), and the large informal sector (figure 2, panel a). The most recent tax gap estimates (2014) suggest that the productivity of revenue from income tax-es is low, particularly in the case of the PIT. Only 12 percent of the PIT tax base was actually taxed, and the corporate income tax rate was similarly low (World Bank 2014). It has been suggested that PIT revenues are especially low in Kosovo because of the PIT exemptions allowed on vari-ous sources of income, weak tax administration, and the size of the informal economy. In addition, a low share of revenue from the PIT relative to the VAT suggests that the system exerts a regres-sive distributional impact, at least on the taxation side, which is particularly the case considering the maximum tax rate is only 10 percent, rather than the more than 40 percent in many EU countries.

Despite improvements in revenue, important

a. Kosovo, by source

b. Western Balkans, total

Figure 2. TAX REVENUE AS A SHARE OF GDP

Source: World Bank calculations based on data from national authorities.

0

2

4

6

8

10

12

14

16

Border VATDomestic VATPITCIT

Shar

e o

f G

DP

2011 2012 2013 2014 2015 2016 2017

0

10

20

30

40

50

20162008

Tax

reve

nu

es (%

GD

P)

Bosnia & Herzegovina

KosovoAlbania FYR Macedonia

Serbia Montenegro

17


challenges persist. Tax revenue as a share of GDP increased from 13 percent to 15 percent between 2014 and 2017; however, the vast majority of this increase arose because of the border VAT (figure 2, panel a). Tax revenues are projected to decline as a share of GDP over the medium term in the absence of measures to bol-ster domestic revenues to compensate for falling border revenues. A gradual shift toward direct taxation will be necessary as domestic produc-tion expands and integration with the EU and free trade agreements advance and thus reduce border revenues (World Bank 2017). Enhanced tax administration is necessary to boost collec-tions and raise compliance.

Information in the tax registry provides a more nuanced profile of compliance in Kosovo. The tax registry, which contains anonymized individ-ual records of taxpayers, including their declara-tion behavior in each tax system, allows for the classification of taxpayers into different compli-ance categories, particularly those who declare or pay late and those who do not submit any tax payments. The tax registry also allows for the classification of firms or individual taxpayers into groups defined by revenue and other measures to distinguish between larger and smaller firms. Assessing the quality of data in the tax registry helps identify major shortcomings hindering the tax collection effort and possible ways to ad-dress them. The PIT registry contains records on 17,696 taxpayers between 2010 and 2016 who declared at least once during this period. The VAT registry contains records on 27,929 firms submitting a monthly declaration at least once

between 2011 and mid-2018. These records in-clude contact information (addresses, e-mails, and telephone numbers) that would be key to any effort to establish a client-oriented tax ad-ministration. Moreover, the registries contain information on corresponding tax centers (ad-ministrative districts), the types of taxpayers in the PIT (natural person or individual business), sectors and types of economic activity (if ap-plicable), the gender of the taxpayer, and other details of the firm or individual business, such as the date the individual or firm went into busi-ness, the date the firm or individual registered with the tax authorities or the business registry, and so on. Evaluating trends in these data, as well as the quality and consistency of the infor-mation, can provide a clearer picture of the fac-tors driving compliance.

Tax registry records in Kosovo demonstrate that individual compliance has improved in recent years; yet, late declarations and late payments are frequent. Figure 3 plots key tax compliance indicators for the PIT and VAT systems, based on tax registry records from 2010 through 2018. Two clear trends emerge: (1) late declarations and late payments are frequent in both systems, but have decreased substantially; and (2) there is substantial fluctuation in compliance behavior among VAT declarations month to month. In the VAT, late payments are endemic among those taxpayers who actually make payments; an av-erage of more than 90 percent of taxpayers sub-mit their payments after the monthly due date.

Strong disparities in tax compliance exist across

18


districts and show signs of persistence, particu-larly in the VAT space. District differences seem to be responsible for most of the heterogeneity in tax compliance, and this is especially true of late declarations—defined as declarations sub-mitted after the official due date—in the PIT space (figure 4). The persistence of district dis-parities is true of the VAT much more than the PIT, where reductions in the share of late dec-larations were largest among districts that, in 2010, were among the worst compliers, such as Ferizaj and Gjakova. In the VAT, compliance in terms of late declarations has either remained high (an average of over 40 percent) or increased in some districts, such as Gjakova, Pristina 2 and 3, and Prizren.

Compliance is strongly correlated with the age of the taxed businesses and the district of ope-ration across both the PIT and the VAT systems. Across both the PIT and the VAT, late decla-rations are a function of key individual or firm characteristics that persist, highlighting the chal-lenges in tax collection that must be addressed. Among individuals shown in the annual PIT re-cords to have submitted declarations since 2010, the share of late declarations is 2 percentage points higher among men than among women. This is also the case among less experienced tax-payers: each additional year an individual or a business has been registered is associated with a 1 percentage point decline in the probability of a late declaration. Taxpayers whose contact information is not recorded in the registry are more likely to submit declarations late; those without an e-mail declare late 47 percent of the

a. PIT declarations

b. VAT declarations

Figure 3. TAX COMPLIANCE TRENDS, BY TAX SYSTEM, 2010–18

Source: World Bank calculations based on data from national authorities. Note: “Late payment” is conditional on the payment having been made, whereas “no payment” refers to both those taxpayers who do not pay the tax burden and those taxpayers who report owing nothing (generally arising because of a credit in their favor from past periods).

0

10

20

30

40

50

60

No paymentLate paymentLate declaration

% o

f d

ecla

rin

g P

IT

taxp

ayer

s in

per

iod

2010 2011 2012 2013 2014 2015 2016

0

25

50

75

100

Late paymentNo paymentLate declaration

% o

f d

ecla

rin

g V

AT

taxp

ayer

s in

per

iod

20

11

m1

20

11

m1

2

20

11

m6

20

12

m6

20

12

m1

2

20

13

m6

20

14

m6

20

14

m1

2

20

15

m6

20

15

m1

2

20

16

m6

20

16

m1

2

20

17

m6

20

17

m5

20

17

m1

2

20

13

m1

2

19


time, compared with those with an e-mail, who are late only 26 percent of the time. Meanwhile, those with no address declare late 41 percent of the time, compared with 27 percent of the time among those whose addresses are on file. Those assumed to be covered by an accountant because of a shared e-mail address are nearly 4 percentage points more likely to declare late relative to those with a unique e-mail address. Taxpayers with a tax burden are 15 percentage points less likely to declare late relative to those without a tax bur-den (16 percent versus 33 percent). Among those with a burden, larger burdens (above the median) are less likely to declare late (15 percent versus 18 percent). Sectoral differences represent an im-portant factor explaining differences in declaring late: relative to those in the primary or second-ary sectors, taxpayers in the tertiary sector are less likely (4 percentage points) to submit their declarations late. The predictive factors associ-ated with late declaration in the VAT are similar to those in the PIT. Younger firms are also more likely to submit their declarations late; each ad-ditional year of appearance on the registry is as-sociated with a 0.6 percentage point decline in the likelihood of late declaration. Firms in the tertiary sector are 3 percentage points less likely to declare late relative to those in the primary and secondary sectors. Those with a tax burden are 13 percentage points less likely to declare late relative to those without a burden, and larg-er firms (above the median of VAT output) are 11 percentage points less likely to declare late. The fluctuations in late declarations by month persist across years. The incidence of late dec-larations is highest in December and lowest in

a. PIT declarations

b. VAT declarations

Figure 4. LATE DECLARATIONS, BY TAX CENTER DISTRICT

Source: World Bank calculations based on data from national authorities.Note: Late declarations are defined as declarations submitted after the due date, officially March 31st of the subsequent year for the PIT and the 20th of the subsequent month for the VAT.

0

10

20

30

40

20162010

% o

f P

IT t

axp

ayer

s in

p

erio

d d

ecla

irin

g la

te

Pri

zren

1

Pri

zren

2

Gjil

an

Pri

stin

a 3

Pri

stin

a 2

Mit

rovi

ca

Gja

kova

Peć

Fer

izaj

Pri

stin

a 1

Avg. 2010 Avg. 2016

0

10

20

30

40

20162010

% o

f P

IT t

axp

ayer

s in

p

erio

d d

ecla

irin

g la

te

Pri

zren

1

Pri

zren

2

Gjil

an

Pri

stin

a 3

Pri

stin

a 2

Mit

rovi

ca

Gja

kova

Peć

Fer

izaj

Pri

stin

a 1

Avg. 2010 Avg. 2016

20


January, April, and July, which highlights the im-portance of using reminders during key periods.

TAK routinely communicates with taxpayers on noncompliance issues, ranging from missing declarations to unpaid or underpaid tax obliga-tions among various types of taxpayers. These communications have taken the form of auto-matic e-mails generated through the electron-ic declaration system, but, more recently, have been dominated by personalized communica-tions through the TAK call center. These calls are less about prevention than about treat-ment: the vast majority of calls are made to taxpayers with substantial outstanding debts or major inconsistencies in declaration behav-ior. Among those who do not declare, there are many exceptions granted for special circum-stances, which are frequently rectified through these communications. Taxpayers in the PIT who earn below a specific threshold (described above) are not required to pay taxes and are thus also not required to declare.6

TAK uses various communication tools to reach out to taxpayers, but, in the past, these tools were not used regularly or evaluated system-atically. This represents a unique opportunity to test the impact of various communication strategies in terms of both the method and the content. TAK classifies noncompliers as taxpay-ers, whether individuals or firms, who have not fulfilled a tax obligation for one or more tax pe-riods and one or more tax systems. Specifically, noncompliance may be categorized as failure to declare (nondeclarers or nondisclosers), failure

to pay (nonpayers), and those taxpayers with outstanding debt (underpayers). The latter two are those who have submitted declarations, but who have not completed their payments. In the-ory, these groups are defined in each tax period immediately after the due date, depending on the system under consideration. At the moment the trials were initiated, TAK did not have any standardized communication strategy in opera-tion to remind taxpayers to declare in advance of the due date. Taxpayers who were late on declarations were generally eligible to receive a phone call from the call center; however, the call itself was not based on a risk assessment. Generally, the prioritization across taxpayers for receiving calls was based on the tax burden and the number of declarations missed and was usu-ally intended to address a specific inconsistency in the records. This means that, effectively, TAK was selective about who to contact, and one may assume that a baseline scenario involved no communication. To guarantee this was the case, during the implementation of the three exper-iments, no communications were undertaken with taxpayers about current declarations. If communications were made, these were related to earlier declarations or a need for some other clarification.

21


3.Behavioral Interventions in Tax Compliance: Three Experiments

Between February and October 2018, the team supported the tax authority in the de-sign, implementation, and evaluation of three experimental trials to improve tax compli-ance in Kosovo.7 The collaboration initially envisioned designing a single trial, but, thanks to an effective use of resources and a fruitful collaboration with TAK, the team was able to implement and evaluate three trials in two tax systems using three different communication strategies. That expectations were exceeded in these trials highlights the adaptive learning potential of randomized experiments in the tax space, that is, the ability to make marginal changes to the design at different stages of an intervention, as well as the benefits of constant local engagement with the tax authority.

The objective of the three tax trials was to evaluate how innovative approaches to com-munication with taxpayers using evidence and adaptive learning could improve timely and honest compliance behavior at low or zero cost. The VAT and the PIT were identified by TAK as priority areas in the effort to improve tax com-pliance, specifically as this related to the number

of tax declarations submitted and their timeli-ness. The PIT space was assigned a priority be-cause of its annual cycle; PIT declarations must be submitted by March 31st each year for the taxes corresponding to the previous calendar year. VAT declarations are required every month, permitting more flexibility around design and continuity. These types of trials, if adequately designed, constitute natural field experiments in the sense that a treatment assignment was randomly allocated to segments of the popula-tion without their knowledge.

The scope of the work and the policy approach revolved around the use of different communi-cation channels and content meant to address the policy challenge, that is, to induce com-pliant taxpayer behavior at low cost without changes to legislation. Communication is only one dimension of many meant to address low tax compliance in Kosovo. However, given the low cost and established communication pro-tocols, this dimension has the highest potential efficiency gains, large expected impacts, and the ability to shed more light on taxpayer behavior by testing different instruments. To identify chang-

22


es in taxpayer behavior that could be directly, that is, causally attributed to the communica-tion channel and content, three experimental trials were developed over the course of eight months. The team used an existing tax sched-ule, declaration infrastructure, and administra-tive data and randomly assigned taxpayers to one of two situations in a given tax system and tax period: those who would be exposed to a business-as-usual scenario (no specific remind-er) and those who would receive one of two re-minders with unique content.8

Each of the three trials involved a slightly dif-ferent taxpayer population.9 Each also involved a different design.10 However, all followed the same proposition: to apply behavioral insights to induce timely, honest declarations among a ran-domly selected subset of the population of in-terest.11 In each trial, the population of interest was divided into three segments: one segment that received no reminder (business as usual); one segment that received a simple, behavioral-ly informed reminder; and one segment that re-ceived the same behaviorally informed remind-er, complemented by additional language that addressed some behavioral bias or biases. The behaviorally informed reminder was written in simple, nontechnical language and contained a clear call to action, some personalization, and in-structions on carrying out the desired behavior. Figure 5 highlights the design aspects of each trial in more detail, and annex figures A.1a–A.4b provide snapshots of the letters and e-mails.12 Given that no standard letter, e-mail, or SMS re-minders were in place at the time of the trial,

all reminders were behaviorally informed be-cause they were designed with the clear intent to make the framing of the messages more con-ducive to the desired behavior.13

To inform the design of the trials, the team car-ried out a behavioral diagnostic. A behavioral di-agnostic may be understood as an assessment of factors that influence behavior in a given space, in this case, the factors that drive individuals toward or away from tax compliance. The diag-nostic was carried out in February 2018. In addi-tion to brainstorming with TAK technical teams, it involved informal discussions with accountant associations, firm owners, and psychologists and semi-structured interviews with individual tax-payers in Pristina and surrounding areas to gauge perceptions of the tax administration and tax compliance in general. These interviews and di-alogues highlighted the perception of corruption and lack of transparency in the use of tax revenue, as well as the perception that larger firms evade taxes. Fieldwork revealed that taxpayers favored the use of tax revenues to finance education and health care. A parallel diagnostic in the jobs space highlights that poor public service delivery may create incentives for informality (Cojocaru 2017). According to Riinvest (2013), over 40 percent of firm representatives believe tax evasion is justi-fied under certain circumstances, including in re-action to the (perceived) unsatisfactory supply of public services.14 Notwithstanding the low levels of trust in public institutions, which is consistent with the findings of national surveys, the percep-tions in regard to TAK were positive, and the lev-els of trust in TAK were relatively high.15

23


Note: The signatures of two department directors were included at the bottom of both letters, and the director of one department at the bottom of the augmented behavioral e-mail.

Not paying your taxes places an unfair

burden on your fellow Kosovo citizens.

Please don’t be an irresponsible citizen.

7 out of 10 firms submit their

declaration on time. Don’t wait,be part of the MAJORITY!

Did you know that your VAT contribution is

invested in your city?

• 7 , 7 4 7 l e t t e r s s e n t

• 2 l e t t e r t y p e s

• 1 1 , 6 0 3 i n d i v i d u a l s l a t e o n 2 0 1 7 P I T d e c l a ra t i o n

• 2 3 , 6 2 2 fi r m s w i t hp e n d i n g d e c l a ra t i o n o n J u n e 2 0 1 8 VAT

• 2 2 , 8 2 8 fi r m s w i t h p e n d i n g d e c l a ra t i o n o n A u g u s t 2 0 1 8 VAT

• 11,096 messages sent

• 2 S M S t y p e s

• 1 0 , 3 4 0 e - m a i l s s e n t

• 2 e - m a i l t y p e s

Previously, we have considered your failure to declare an oversight. However, if you do not

declare now, we will consider it to be your

active choice…

Figure 5. DESIGN CHARACTERISTICS OF THE TAX TRIALS

The first trial targeted letter reminders at PIT taxpayers who were late in filing their 2017 annual declarations. Taxpayers who, by April 4, 2018, four days after the due date, had not yet submitted their PIT declarations for the 2017 fiscal year were identified in the tax registry and were eligible to receive a let-ter in the mail.16 These letters were posted by private courier to record those who did not re-ceive the letter because of incorrect or outdated

addresses, which provided valuable information on the efficiency of the tax registry. The eligible taxpayers (11,603 in total) were divided ran-domly into one of three groups of similar size: (1) those who would receive no letter; (2) those who would receive a simple, behaviorally informed let-ter; and (3) those who would receive the same let-ter as group 2, but augmented with an additional behaviorally informed message. The letters sent to group 3 included deliberate choice language,

24


which emphasized that a failure to act would be considered a choice of the taxpayer rather than an oversight. The private courier attempted to deliv-er a total of 7,747 letters; multiple attempts were made to locate proper addresses and deliver the letters. A recipient signature was requested, and the details of the delivery or the attempt to deliv-er were recorded by the courier. The letters were intended to be delivered within three days of the assignment of the recipients into the three groups.

The second trial targeted all firms required to submit June VAT declarations; it involved the use of e-mail–based reminders on two occa-sions, the first well in advance of the monthly due date. All firms in the VAT registry, excluding those that had already submitted declarations by July 3, 2018, or for which no e-mail addresses were on file, were considered eligible to receive an e-mail reminder from the TAK Information Technology Department, first on July 4, then again on July 20, the due date of the June declaration. The eligible firms (23,622 in total) were divided randomly into one of three groups of similar size: (1) those who would receive no e-mail; (2) those who would re-ceive a simple, behaviorally informed e-mail; and (3) those who would receive the same e-mail as group 2, but with reference to the unique fiscal numbers of the corresponding taxpayers and augmented with positively framed, behaviorally informed messag-es placed strategically throughout the body of the e-mail.17 These short behavioral messages includ-ed (1) language to instill feelings of guilt for non-compliance, (2) language highlighting the benefits that taxes provide (the public goods message), and (3) language highlighting how the majority of other

taxpayers fulfill their obligations (the social norms message). For the VAT e-mail trial, the TAK Informa-tion Technology Department sent out the messages using their own software. All e-mails were intended to be sent within the same time frame, a single day for both the first and the second rounds of e-mails. E-mails were sent in the second round regardless of whether the first e-mail was opened or the taxpay-er had already submitted a declaration.

The third trial targeted all firms required to sub-mit August VAT declarations; it involved the use of SMS-based reminders a few days in advance of the due date. All firms in the VAT registry, excluding those that had already submitted their declarations by September 17, 2018, or for which there was no cell phone number on file, were considered eligible to receive an SMS reminder, which was sent through a telecom-munication firm in Kosovo. The eligible firms (22,828 in total) were divided randomly into one of three groups: (1) those that would re-ceive no SMS message; (2) those that would receive a simple, behaviorally informed SMS message; and (3) those that would receive the same SMS message as group 2, but augmented with a social norms message.18

“Dear taxpayer, please declare and pay your August VAT correctly and on time!,” read the simple reminder, and the social norms language added “7 of 10 firms submit their declarations on time. Don’t wait, be part of the MAJORITY!”19 A telecommunication company was hired to send all the SMS messages during a one-hour window on September 18, two days before the due date.

25


All three behaviorally informed communication trials succeeded in producing statistically sig-nificant impacts on the declaration rate or the payment rate in the short term, within four to six weeks. The behavioral messages managed not only to induce more people to submit their declarations on time, but also to encourage more people to submit declarations.20 These impacts hold under various specifications, includ-ing limiting samples to specific subgroups among which impacts may be expected to be low.21 This highlights the value of testing, learning, and adapt-ing in tax interventions. The result is all the more encouraging considering the low take-up rate in two of the three trials. In those two trials, many letters did not reach the intended recipients, and e-mails were often not read. Moreover, the total amount of taxes declared or paid did not rise in this first, limited exercise, partly because many of the intervention recipients were not obligat-ed to submit declarations. The short-term objec-tive of these trials was to increase the number of individuals and firms that submitted timely declarations. Declaration is the first step toward increasing tax revenues because it builds a so-cial norm around compliance and the social con-

tract (the shared principles that serve to regulate, define, and redistribute national resources). Over-all, the effectiveness of the trials can be sum-marized by answering the following questions on each trial:

• Did the various reminders directed at tax-payers encourage them to declare and pay in a timely fashion?

• Did the message content lead to a separate change in behavior among taxpayers that was associated with each communication method?

• Did taxpayers respond differently to the reminders and the content of the messages?

The results of each of the three trials are pre-sented below. Many of the people who should have received reminders did not receive them be-cause of the incorrect or outdated postal address-es or because they ignored the e-mails. The asso-ciated intervention effects are therefore examined on the basis of those who should have received the reminders (intent-to-treat) and those who actually did receive the reminders (average treat-

4.Summary of Results

26


ment-on-the-treated). (Most SMS reminders were received and viewed by the intended recipients.) Had more addresses been correct or had more people opened e-mails, the general results would have been better because they would have cor-responded to the greater incidence of effects ob-served on those who did receive the reminders.

4.1Letter Reminders to Taxpayers Late in Submitting their PIT Declarations

Letter reminders to PIT taxpayers late in sub-mitting their annual declarations resulted in small, but statistically significant impacts on the declaration rate. The reminders were successful in inducing more people to submit declarations closer to the due date. Among those who were randomly assigned one of the two letters (the intent-to-treat group), there was an increase of 2 to 4 percentage points in the declaration rate within one month of the transmission of the let-ters.22 Only half the people assigned to receive a letter did receive one. This shortfall was caused by the large number of incorrect addresses in the records. If only the group is examined that actually received the letters (the treatment-on-the-treated estimate), the increase in the decla-ration rate is almost double, at 4 to 7 percent-age points.23 This represents a large increase in

the declaration rate among late declarers, which increased by between 91 and 160 percent (fig-ure 6). The difference in the magnitude of effects on both groups—those assigned to receive a let-ter and those who actually did receive one—is important. This is because, in the first case, the overall effect is analyzed in the trial, whereas, in the second case, the effect on the treated group is analyzed, and this indicates what might have been the overall outcome if the take-up problem had been dealt with by correcting and updating postal addresses, for example.

The trial had different effects across the groups. The heterogeneity in the effects on the declaration rate was appreciable; im-pacts were substantially larger among some groups. This was particularly marked across the districts in which the taxpayers resided and ac-cording to the past declaration behavior of the taxpayers. For example, in relative terms, the im-pact was considerably larger among natural per-sons than among individual businesses. While the share of natural persons who had submitted dec-larations on time rose three- to fourfold (between 270 and 431 percent increase), the correspond-ing share of individual businesses was more mod-est, but still substantial, between 48 percent and 94 percent increase. However, the latter group reported a much higher rate of declaration in the absence of treatment than the former (11 percent versus 1 percent).

Despite the significant impacts on the dec-laration rate and the payment rate, no ef-

27


Source: World Bank calculations based on data from national authorities.Significance level: *** = 1 percent. ** = 5 percent Bars with arrows represent confidence intervals.

c. Declaration rate, by group, March–July 2018

a. Letter assigned b. Letter received

No reminder Simple reminder +deliberate choice

Simple reminder0

4

8

12

16

20

PIT

dec

lara

tio

n r

ate

(%) a

mo

ng

tho

se f

aili

ng

to d

ecla

re b

y d

ue

dat

e

44%73%

No reminder Simple reminder+deliberate choice

Simple reminder0

4

8

12

16

20

PIT

dec

lara

tio

n r

ate

(%) a

mo

ng

tho

se f

aili

ng

to d

ecla

re b

y d

ue

dat

e

91%

160%

******

***

***

No reminder Simple reminder +deliberate choice

Simple reminder0

4

8

12

16

20

PIT

dec

lara

tio

n r

ate

(%) a

mo

ng

tho

se f

aili

ng

to d

ecla

re b

y d

ue

dat

e

44%73%

No reminder Simple reminder+deliberate choice

Simple reminder0

4

8

12

16

20

PIT

dec

lara

tio

n r

ate

(%) a

mo

ng

tho

se f

aili

ng

to d

ecla

re b

y d

ue

dat

e

91%

160%

******

***

***

0

2

4

6

8

10

Simple reminder + deliberate choiceSimple reminderNo reminder

PIT

dec

lara

tio

n r

ate

(%) a

mo

ng

tho

se f

aili

ng

to d

ecla

re b

y d

ue

dat

e

28

-Mar

01

-Ap

r0

5-A

pr

09

-Ap

r1

3-A

pr

17

-Ap

r2

1-A

pr

25

-Ap

r2

9-A

pr

03

-May

07

-May

11

-May

15

-May

19

-May

23

-May

27

-May

31

-May

04

-Ju

n0

8-J

un

12

-Ju

n1

6-J

un

20

-Ju

n2

4-J

un

28

-Ju

n0

2-J

ul

06

-Ju

l1

0-J

ul

Figure 6. SHORT-TERM IMPACT OF LETTERS ON PIT DECLARATIONS AMONG TAXPAYERS FAILING TO DECLARE BY THE DUE DATE, 2017

28


fects were identified on payment amounts. While the rate at which taxpayers declared and made a payment was substantially higher among those receiving a letter (1 to 2 percentage points higher), the amounts paid were not different among those who received a letter. This is not surprising, however, because the population of interest in this trial plausibly consisted of indi-viduals who did not actually have anything to declare. Indeed, among those who were late in submitting declarations, but who did end up submitting in the absence of the reminder, only 18 percent claimed they owed something in tax-es. Conditional on owing something, those who received a letter did, however, submit a nonzero payment at a higher frequency relative to those who did not receive a letter. In other words, among those who owed taxes and submitted declarations, the group that received reminders was larger than the group that did not receive reminders.

The simple behavioral letter was as effective as the behavioral letter augmented by a deliberate choice message. This outcome is not predicted by theory. Relevant experience in other coun-tries suggests that the inclusion of the deliberate choice message, along with the simple reminder, tends to be significantly more effective than the simple reminder alone. The reason is that the ad-ditional message highlights that it is an individu-al’s choice to fail to submit a declaration and that failure to submit is not merely an unintentional error. However, in the trial, the magnitude of the impact was larger in the case of the simple letter compared with the augmented letter. This

outcome may derive from the way the deliberate choice message was framed in the augmented letter. The phrasing might have been interpreted by some recipients as a veiled threat. Most like-ly, however, the unexpectedly low impact of the deliberate choice message may be the result of operational difficulties during implementation.

4.2 E-mail Reminders to Firms Subject to the Value Added Tax

The e-mail reminders sent to firms in July did not lead to a substantial increase in the submission of declarations for June taxes, in payment rates, or in the amount of taxes paid.24 The e-mail re-minders led to an average 0.5 to 1.2 percentage point rise in the on-time declaration rate and an average 1.0 to 1.6 percentage point rise in the declaration rate within two weeks after the due date, though neither of these changes was statis-tically significant. A small share, 25 percent, of the e-mails received by firms were actually opened. One might expect that the firms at which e-mails were opened would show a larger effect. Yet, this was not entirely the case (figure 7). Firms at which e-mails were received and opened did submit dec-larations on time more often than the other firms, but this could not be captured in a statistically sig-nificant manner because the number of firms at which e-mails were opened was so small. Similar results were obtained in the case of the positively

29


framed e-mails. The declaration rates were higher over time among firms at which positively framed e-mails were received and opened, but the share of these firms was so small that the results were not statistically significant, and causality cannot be claimed with high confidence. The e-mails including the additional positive framing were opened by only 7 percent of the firms receiving them, whereas the simple e-mails were opened by half the recipient firms. Among the 7 percent of firms at which the e-mails with the augmented positive framing were opened, 74 percent submitted declarations within two weeks following the due date, compared with 67 percent among firms at which the e-mails were not opened. This difference was substantially small-er—67 percent among firms at which the e-mails were not opened and 70 percent among firms at which the e-mails were opened—among firms re-ceiving the simpler e-mail reminder.

The augmented behavioral message did, howev-er, have a positive impact on the share of taxpay-ing firms submitting a declaration and making a payment within a month and a half after the due date. Conditional on submitting a declaration by early September, those firms that received the e-mail with a positive framing of the benefits of tax-es were 2.2 percentage points more likely to make a payment (figure 8). This is an important finding because payment is generally made substantially later than the declaration and rarely by the due date. Conditional on opening the e-mail reminder with augmented behavioral language, this impact increases to a remarkable 32 percentage points. Notwithstanding the rise in the payment rate, no increases were observed in the payment amounts

Figure 7. DECLARATION RATE, BY E-MAIL ASSIGNMENT AND STATUS OF RECEPTION

a. By asignment to rceive the e-mail

0

20

40

60

80

Simple reminder + positive framingSimple reminderNo reminder

VA

T d

ecla

rati

on

rat

e (%

), J

un

e

04

-Ju

l0

6-J

ul

08

-Ju

l1

0-J

ul

12

-Ju

l1

4-J

ul

16

-Ju

l1

8-J

ul

20

-Ju

l2

2-J

ul

24

-Ju

l2

6-J

ul

28

-Ju

l3

0-J

ul

01

-Au

g0

3-A

ug

05

-Au

g0

7-A

ug

09

-Au

g1

1-A

ug

13

-Au

g1

5-A

ug

17

-Au

g1

9-A

ug

b. Among those for which e-mails were opened


0

20

40

60

80


VA

T d

ecla

rati

on

rat

e (%

), J

un

e

04

-Ju

l0

6-J

ul

08

-Ju

l1

0-J

ul

12

-Ju

l1

4-J

ul

16

-Ju

l1

8-J

ul

20

-Ju

l2

2-J

ul

24

-Ju

l2

6-J

ul

28

-Ju

l3

0-J

ul

01

-Au

g0

3-A

ug

05

-Au

g0

7-A

ug

09

-Au

g1

1-A

ug

13

-Au

g1

5-A

ug

17

-Au

g1

9-A

ug

30


as a result of either of the two e-mails.

E-mail reminders had different effects across taxpayer subgroups; more analysis is needed to understand this heterogeneity. Reminders had substantial positive impacts on the declara-tion rate among firms below the median of VAT output, that is, the amount of VAT charged to consumers, and among firms that had recent-ly failed to submit declarations, for example, in the previous few months. However, the ef-fect of the e-mail assignment was negative in the case of some groups. More analysis of the tax registry data is needed to understand these

Figure 8. PAYMENT RATE, BY E-MAIL ASSIGNMENT AMONG FIRMS THAT SUBMITTED DECLARATIONS


0

4

8

12

16

20


VA

T p

aym

ent

rate

(%),

Ju

ne

04

-Ju

l0

6-J

ul

08

-Ju

l1

0-J

ul

12

-Ju

l1

4-J

ul

16

-Ju

l1

8-J

ul

20

-Ju

l2

2-J

ul

24

-Ju

l2

6-J

ul

28

-Ju

l3

0-J

ul

01

-Au

g0

3-A

ug

05

-Au

g0

7-A

ug

09

-Au

g1

1-A

ug

13

-Au

g1

5-A

ug

17

-Au

g1

9-A

ug

differential impacts.

4.3

SMS Reminders to Firms Subject to the Value Added Tax

SMS reminders sent before the due date led to a 2.1 to 3.7 percentage point increase in the on-time declaration rate and a 2.2 to 3.9 percent-age point increase in the declaration rate (figure 9). Within one month of the due date, both SMS variants led to a statistically significant rise in the declaration rate. Given the high level of treat-ment compliance (80 percent of the messages were received by taxpayers), it is not surprising that these results are practically the same if they are conditioned on actually receiving the SMS messages. Payment rates (the likelihood that a firm submitted any payment) were not high-er among firms that received an SMS message; so, while the reminders induced firms to declare sooner, they did not induce them to submit pay-ments sooner.

Similar to the letter and e-mail trials, there was no impact on payment amounts in the SMS trial. Payment amounts were no higher among firms that received the SMS reminders than among those that received no reminder. This outcome may have arisen because of the tendency to make payments a substantial time after submitting the

31


Figure 9. IMPACT OF SMS MESSAGES ON VAT DECLARATIONS AFTER ONE MONTH

a. Letter assigned b. Letter received

50

55

60

65

70

75

VA

T d

ecl

ara

tio

n r

ate

(%

), A

ug

ust

No reminder Simple reminder +social norm

Simple reminder50

55

60

65

70

75

VA

T d

ecl

ara

tio

n r

ate

(%

), A

ug

ust


Simple reminder

3%**

5%***

6%***4%

**

50

55

60

65

70

75

VA

T d

ecl

ara

tio

n r

ate

(%

), A

ug

ust


Simple reminder50

55

60

65

70

75

VA

T d

ecl

ara

tio

n r

ate

(%

), A

ug

ust


Simple reminder

3%**

5%***

6%***4%

**

c. Declaration rate, by treatment group

0

20

40

60

80

Simple reminder + social normSimple reminderNo reminder

VA

T d

ecla

rati

on

rat

e (%

), A

ugu

st

16

-Sep

18

-Sep

20

-Sep

22

-Sep

24

-Sep

26

-Sep

28

-Sep

30

-Sep

02

-Oct

04

-Oct

06

-Oct

08

-Oct

10

-Oct

12

-Oct

14

-Oct

16

-Oct

18

-Oct

20

-Oct

Source: World Bank calculations based on data from national authorities.Significance level: ** = 5 percent, *** = 1 percent. Bars with arrows represent confidence intervals.

32


declaration, or it may simply reflect the fact that the messages served more as nudges to do some thing sooner, but not necessarily better. A simple follow-up with the SMS message recip-ients would reveal if there is an effect on pay-ments or a longer-term effect on declaration rates overall and on on-time payments.

Heterogeneity in the impacts of the SMS re-minders was found among select subgroups of taxpayers, though more analysis is needed to understand this heterogeneity. The effects of the SMS reminders on the declaration rate were particularly marked among firms in the pri-mary and secondary sectors. Thus, there was a 5 percentage point rise in the declaration rate associated with the simple SMS message. More analysis of the tax registry data is needed to un-derstand these differential impacts.

33

Kosovo Promoting Tax Compliance with Behavioral Insights

A crucial component of effective experimental trials to improve tax compliance is an under-standing of the implementation and capaci-ty-building process. The effectiveness of an in-tervention is dependent, after all, on successful implementation. An analysis of this process is perhaps even more important than the evalu-ation of impacts because it uncovers the risks and challenges faced in tax compliance in gen-eral, which are undoubtedly associated with behavioral factors that may not be revealed and addressed through the exploration of impacts alone. The approach to designing the behav-ioral trials followed the process illustrated in figure 10.

The pretrial diagnostic revealed that a taxpayer as client model is an effective way to improve perceptions of the tax administration. Semi-structured interviews with taxpayers and in-formal dialogue with key stakeholders revealed that attitudes toward TAK are generally positive, highlighting the voluntary nature of tax collec-tion that the institution values.25 Some firm owners and accountants indicated that small-er, less accomplished firms are pursued with

more frequency than larger, more well estab-lished firms, and this was related to the belief that many firms underreport or altogether evade the taxes that they owe. This is also highlighted in the taxpayer survey conducted in Kosovo in 2017 (UBO Consulting 2018).

Brainstorming with TAK helped support this narrative and direct the message content to-ward positively framed statements. Insights communicated by the TAK teams supported the importance of positively framed messages, specifically, messages that placed less empha-sis on enforcement and more on positive rein-forcement by highlighting the benefits that tax-es provide and targeting the moral inclinations of individuals. The capabilities and efforts of TAK greatly facilitated the trial process, espe-cially with regard to existing data systems and processes that allowed for a baseline to be es-tablished and the process of treatment assign-ment in the PIT and VAT trials to be simplified.

5.Process Makes Perfect: Lessons Learned from Implementation

34


5.1

Implementation Successes

The partnership with the GIZ was motivat-ed by the recognition that this type of pol-icy challenge could most effectively be ad-dressed through a collaboration embracing the strengths of each institution in the field. Many implementation successes were achieved through the three trials, both in the trials them-selves and in the overall coordination among the World Bank, the GIZ, and jointly with TAK. Con-sidering the positive impacts, these successes

demonstrate the benefit of using behavioral science approaches to public policy. The World Bank was able to provide expertise in solutions to implementation and evaluation problems based on evidence associated with behavioral insights, while advancing its current engage-ment with the Ministry of Finance to improve tax collection in general. The GIZ was strate-gically positioned to buttress implementation and effectively manage the coordination ef-fort with the tax administration, given the GIZ-supported Public Finance Project team’s long-standing relationship and continued en-gagement in capacity building with the tax administration. This was the first partnership between the World Bank team and the GIZ in

01.

Definition

& Diagnosis

03.

Implementation

& Evaluation

04.

Adapt

05.

Re-define

&

Re-diagnose

02.

Design

Figure 10. APPROACH TO A BEHAVIORALLY INFORMED TRIAL

35


Kosovo, and the initiative provided important lessons for future engagement between the two institutions in Kosovo and beyond.

Coordination between the GIZ and the World Bank proved that a partnership embracing the strengths of each team can capitalize on these strengths and exploit synergies to achieve successful results. The GIZ acted as a liaison with TAK on day-to-day aspects of the trials, maintaining an open line of communication with leadership and technical teams and tak-ing advantage of an existing dialogue with the administration. Moreover, the GIZ played a piv-otal role in supporting a behavioral diagnostic to inform the design of the trials, including car-rying out semistructured interviews alongside members of the extended World Bank team and following up with multiple stakeholders. While initial differences in the understanding of the division of labor among the two teams needed to be addressed, including the communication of objectives and the ways to reach them, the insistence by each team on investing in the bi-lateral dialogue proved invaluable in the suc-cess of the project. The availability of a team on the ground that has a close relationship with the client, while incorporating frequent feed-back from multiple stakeholders, is key to reaching intended outcomes.

Close coordination with TAK allowed for quick action and feedback and demonstrat-ed that a behaviorally informed nudge might encourage a population of interest toward a common goal, but also create positive exter-

nalities by nudging the policy makers. The team was able to coordinate the content of messages and communication channels with the TAK director general and his team, which (1) guaranteed that the trials were informed by the knowledge and preferences of TAK and (2) allowed TAK to witness the design of a behav-iorally informed intervention from the ground up and how adjustments could be made to re-flect the different realities of taxpayers. By par-ticipating in the design and implementation of the trials through brainstorming sessions with the TAK Director General’s Office and the TAK Information Technology Department, TAK staff were able to come to understand how behav-ioral science tools can change the mindsets of taxpayers and tax administrators by highlight-ing nuances in the communication with taxpay-ers. The coordination with TAK also underlined how TAK data systems could be pushed to the limit to identify new ways of doing business and interacting with taxpayers.

Considering the unique implementation chal-lenges that arose, the engagement was a remarkable success in building capacity, improv-ing future policy, and establishing important collaborations. The GIZ was able to coordinate successfully with two firms to deliver letters across the country and to send SMS messages even when information was either erroneous or incomplete. The GIZ’s frequent communication with both firms on the delivery of messages and recording observations during the process made the team’s evaluation of information on the success of the trials substantially easier.

36


5.2Implementation Challenges

Despite the successes of the three trials, im-plementation challenges created bottlenecks in the design and, ultimately, the evaluation of the trials; these challenges should be ad-dressed in future trials in Kosovo and else-where. These challenges were observed in each of the trials rather than in the coordina-tion across the World Bank, the GIZ, and the TAK teams. The challenges faced during each trial are briefly described below.

The implementation challenge of the PIT letter trial centered on the lack of reliable contact information for the delivery of messages to taxpayers. Among the population of interest, that is, taxpayers who, by the due date, had not submitted tax declarations, the postal ad-dresses of 27 percent were not on file or were outdated or erroneous and could not be used in the trial. Among the remaining 11,600 tax-payers with an address on file, only 48 percent of the 7,747 taxpayers randomly selected to receive a letter could be found despite multi-ple attempts by the courier service, including phone calls to the taxpayers to validate contact information. This revealed an unfortunate issue with the tax registry: the quality of informa-tion on addresses is poor, including addresses that were ultimately located by the courier. In addition, less than two-thirds of those who re-

ceived a letter actually signed for it, suggesting that the intended recipient did not necessarily receive or even read the letter (figure 11).

Moreover, because of operational difficulties, the timing of letter delivery was different for each of the treatment arms, which is not ideal. The delivery of the standard letter started one full day before the delivery of the deliberate choice letter. The deliberate choice letter arrived an average of more than a half day later and clos-er to the due date. Nearly half the standard let-ters were delivered by the fifth day, whereas only a third of the deliberate choice letters were deliv-ered by the same date. This is relevant because the letter urged action by a specific date, and the later the letter was received, the more likely a tax-payer would presumably give up on the process, believing that there was no time to act.

The implementation of the VAT e-mail trial also faced a number of technical challenges. While a larger share of VAT taxpayers had an e-mail ad-dress on file relative to the postal addresses in the PIT trial, around 20 percent of the e-mails origi-nally sent out by the TAK Information Technolo-gy Department bounced back because the e-mail address was not valid. This may have occurred because the e-mail addresses were no longer ac-tive or were incorrectly entered into the registry by the taxpayers. This problem of treatment non-compliance was exacerbated by the fact that few-er than 25 percent of the e-mails that arrived at valid e-mail addresses were opened, despite two attempts at different dates, that is, at the begin-ning of the month and around the due date, 15–20

37


days later. The rate at which e-mails were opened was substantially higher in the case of the stan-dard e-mail (more than 50 percent) relative to the positively framed e-mail (only 7 percent). Many of the e-mails may not have been opened because of issues in the delivery. The first e-mail remind-er included inconsistent subject lines across the two variants, as well as unbalanced timing: some e-mails were sent early in the morning; others late in the evening or overnight. That they were opened with much less frequency is not surpris-ing. The positively framed e-mail was also sent an average of more than a half day later. The second e-mail reminder was also poorly timed. Moreover, a glitch in the batch e-mail delivery system re-sulted in a delay, causing half the e-mails to be sent six days after the due date. The batch e-mail procedure required a member of the information technology team to send each batch individually, which meant that some language and treatment group combinations were sent at more opportune times than others (figure 12). Independent of im-plementation issues, factors that drive an individ-ual’s decision to open messages were identified in both rounds of e-mail delivery. The most im-portant determinant in whether or not an e-mail was opened was the treatment assignment, which was highly correlated with the time the e-mail was sent. Those receiving the extended behav-ioral e-mail were around a third as likely to open the message as those receiving the standard be-havioral message, after one controls for the hour during which the e-mail was sent. Another import-ant factor was the number of years a firm has been registered. Accountants representing longer-ten-ured firms were less likely to open the e-mail.

The implementation of the VAT SMS trial, while relatively more successful in terms of enhancing compliance than the letters or the e-mail reminders, was nonetheless hindered by issues related to data quality. The SMS trial displayed the highest exposure of all three trials: 80 percent of all SMS messages were success-fully delivered to the phone numbers on file, though a small share of the phone numbers were invalid. While the telecommunication company sending the SMS messages had no way of know-ing if the messages were read, it is fair to assume that the reminder was available on the phones because most phones allow the first portion of a message to be displayed even if the full message is not opened. (The full message was limited to 160 characters.) Timing was largely fixed across all treatment arms. All SMS messages were sent between 10:00 a.m. and 11:00 a.m. on the same day. This was not the case with e-mails; so, all taxpayers who received an SMS message did so at a moment when they might be more likely to read it, that is, earlier during the workday. De-spite these positive highlights, multiple phone numbers exist for each taxpayer, and simple rules had to be established to identify a pri-mary cell phone number, which did not always result in a correct phone number and possibly meant these messages were sent to the wrong person. Moreover, the quality of the numbers meant that some could not be used at all in the trial (some had the incorrect number of digits, for example); for others, assumptions had to be made on leading 0s, country codes, and so on to identify a valid phone number.

38


5.3Opportunities Ahead

Determining the most effective communication channel is context specific and may not align with the priorities of a tax administration. For example, e-mail was identified as the most effec-tive method of communication with taxpayers, but the majority of the e-mails were not read by the recipients. SMS messages were more suc-cessful in delivery, opening, and reading rates. Given cultural considerations, SMS messages might be the most effective in future trials simply because of the reliability of the contact informa-tion and despite the low impact of the messages on declaration behavior. The prominence of tax accountants in helping individual taxpayers and firms submit their declarations is also an import-ant factor to consider because the intended audi-ence might not be receiving the message, which was likely the case in many e-mail and some SMS communications.

Existing communication strategies and the method of communication employed might not be optimal. Consider, for example, the use of e-mails to increase VAT tax compliance. Only one e-mail in five was actually opened. In the future, test e-mails might be sent to help assess whether the e-mail addresses are val-id and whether the e-mails are opened and to ask taxpayers to complete a short question-naire on their favored method of communica-tion. Enhancing the quality of the data on postal

Figure 11. DELIVERY OF LETTERS


0

10

20

30

40

50

60

70

Address found

% o

f ad

dre

sses

fo

un

d

Nu

mb

er of P

IT letters p

repared

Fer

izaj

Mit

rovi

ca

Gja

kova

Pri

stin

a

Gjil

an

Pri

zren Peć

Number of letters

0

800

1600

2400

3200

4000

0

25

50

75

100

Simple reminder + deliberate choiceSimple reminder

% o

f al

l PIT

lett

ers

del

iver

ed

06

-Ap

r

07

-Ap

r

08

-Ap

r

09

-Ap

r

10

-Ap

r

11

-Ap

r

12

-Ap

r

13

-Ap

r

a. By district

b. By delivery date

39


addresses, e-mail addresses, and phone num-bers is fundamental to improving communica-tion with taxpayers.

Continuing the dialogue in Kosovo has potential in establishing a testing ground for the realization of advances in the development and implementa-tion of communication strategies. Two methods of communication that were not tested could be considered in future trials. Throughout the project, it became clear that the more personal or individu-alized the communication, the more likely a nudge would work. Individualized phone calls from the TAK call center might provide a useful, albeit re-source-dependent means to test behavioral mes-sages among taxpayers. TAK already reaches out to select taxpayers by phone to inquire about late declarations and late payments; so, a trial could take advantage of the existing infrastructure and test different scripts. Website-based messaging through pop-ups or the use of CAPTCHAs, for ex-ample, could also be effective. This would entail building a communication infrastructure within the declarations platform and testing how vari-ous message formulations change declaration and payment behavior among taxpayers who log in to the website (see Kettle et al. 2017). Such a com-munication strategy would be low cost; most of the cost would be incurred in establishing, rather than maintaining the infrastructure, which can be tweaked at low marginal cost.

Message content and wording are also con-text specific. Negatively toned messages, which were successful in a similar trial in Poland, seem

to elicit an undesirable response from taxpayers in Kosovo, evidenced by complaints communi-cated to call centers, even though the intention was to keep messages more positively framed, as was the case with the deliberate choice lan-guage. Moreover, the behaviorally augmented messages (beyond using simplified language and a clear call to action) did not show substantial-ly greater impact than simple reminders, sug-gesting that messages need to be tested before implementation at scale. Future interventions might involve focus groups to test various mes-sages and inform message content.

The support of local partners is indispensable for successful implementation and long-term capacity building. Supporting the implementa-tion of tax trials requires a hands-on approach by the project team, which is feasible only with local support. Because decisions are generally made in a tight time frame and data are shared on the fly, daily communication with the tax au-thority during implementation is crucial for suc-cess in these types of interventions. By closely collaborating during the trials, the World Bank and the GIZ were able to foresee the challeng-es posed by the project and make quick adjust-ments if needed. Ongoing cooperation between these two institutions is essential. Partners can learn from this implementation success, especially if the collaboration with a govern-ment counterpart is communication intensive. Close collaboration between the teams creates mutual learning and effective adaptation, in-stilling a behavioral mindset among the GIZ and

40


Figure 12. TIMING OF E-MAIL DELIVERY

a. First round of e-mails (early July) b. Second round of e-mails (late July)

04

-Ju

l 12

:00

04

-Ju

l 13

:00

04

-Ju

l 14

:00

04

-Ju

l 15

:00

04

-Ju

l 16

:00

04

-Ju

l 17

:00

04

-Ju

l 18

:00

04

-Ju

l 19

:00

04

-Ju

l 20

:00

04

-Ju

l 21

:00

04

-Ju

l 22

:00

04

-Ju

l 23

:00

05

-Ju

l 00

:00

05

-Ju

l 01

:00

05

-Ju

l 02

:00

05

-Ju

l 03

:00

05

-Ju

l 04

:00

05

-Ju

l 05

:00

05

-Ju

l 06

:00

05

-Ju

l 07

:00

05

-Ju

l 08

:00

05

-Ju

l 09

:00

05

-Ju

l 10

:00

05

-Ju

l 11

:00

05

-Ju

l 12

:00

05

-Ju

l 13

:00

05

-Ju

l 14

:00

05

-Ju

l 15

:00

05

-Ju

l 16

:00

05

-Ju

l 17

:00

05

-Ju

l 18

:00

05

-Ju

l 19

:000.0

0.2

0.4

0.6

0.8

1.0

Simple reminder + positive framingSimple reminder

Shar

e o

f e−

mai

ls d

eliv

ered

0.0

0.2

0.4

0.6

0.8

1.0

Simple reminder + positive framingSimple reminder

Shar

e o

f e−

mai

ls d

eliv

ered

20

Ju

l 08

:00

20

Ju

l 18

:00

21

Ju

l 04

:00

21

Ju

l 14

:00

22

Ju

l 00

:00

22

Ju

l 10

:00

22

Ju

l 20

:00

23

Ju

l 06

:00

23

Ju

l 16

:00

24

Ju

l 02

:00

24

Ju

l 12

:00

24

Ju

l 22

:00

25

Ju

l 08

:00

25

Ju

l 18

:00

26

Ju

l 04

:00

26

Ju

l 14

:00

27

Ju

l 00

:00

c. Sent vs. opened e-mails, first round of e-mails

0

100

200

300

400

500

600

700

800

900

OpenedSent

# E

mai

ls

0:0

01

:00

2:0

03

:00

4:0

05

:00

6:0

0

9:0

01

0:0

01

1:0

01

2:0

01

3:0

01

4:0

01

5:0

01

6:0

0

17

:00

18

:00

19

:00

20

:00

21

:00

22

:00

23

:00

7:0

08

:00


41


TAK partners and ensuring that future policies and programs incorporate behavioral design at the outset.

The bigger challenge—the elephant in the room—in tax administration is the need to address the informal economy. Using the evi-dence-based insights outlined in this brief, fu-ture efforts should also be directed at expand-ing the size of the tax registry by integrating informal income earners and businesses in the system. Simple messages can be effective at en-couraging individuals and firms to do their part by registering with the tax and business regis-tries. By using the existing data infrastructure through the census or public service providers, similar strategies could expand the size of the tax base and address the more serious fiscal challenges in Kosovo.

42


6.ConclusionInstitutionalizing a behavioral approach to tax collection could enable Kosovo to develop in-terventions that are effective locally at low cost and with only modest demands on administra-tive and human resources. The potential positive impact of behavioral interventions more than jus-tifies the cost of experimentation. Establishing a team that conducts rigorous evaluations would not only help identify what works, but also enable policy makers to understand taxpayer behavior. Determining whether interventions create per-sistent and lasting effects among taxpayers is important. The effectiveness of behavioral inter-ventions may diminish over time; thus, remind-er messages may become routine, and the per-ceived threat of enforcement may fade. Adapting communication strategies by changing the con-tent of messages and contacting taxpayers in dif-ferent ways may signal the continued attention of the authorities, and fostering this perception may promote taxpayer compliance independent-ly of the messages.

Technological advances are reducing the already low cost and limited administrative demands of the application of behavioral insights to tax col-lection and public policy more generally. Tax au-thorities could leverage existing technologies such as websites or smartphones to promote compliance with modest budgetary resources. These types of experiments also produce gran-ular data on taxpayers, which can inform pol-icies. Thus, building institutional capacity for experimentation has the potential to generate large returns at comparatively low cost, espe-cially in lower-middle-income countries such as Kosovo.

43


References

A

Arcos Holzinger, Lilia, and Nicholas Biddle. 2016. “Behavioural Insights of Tax Compli-ance: An Overview of Recent Conceptual and Empirical Approaches.” TTPI Working Paper 8/2016 (October), Tax and Transfer Policy Institute, Crawford School of Public Policy, College of Asia and the Pacific, Australian National University, Canberra.

Ariel, Barak. 2012. “Deterrence and Moral Persuasion Effects on Corporate Tax Com-pliance: Findings From A Randomized Con-trolled Trial.” Criminology 50 (1): 27–69.

B

Brockmeyer, Anne, Marco Hernandez, Stewart Kettle, and Spencer Smith. Forth-coming. “Casting a Wider Tax Net: Experi-mental Evidence from Costa Rica.” American Economic Journal: Economic Policy. https://www.aeaweb.org/articles?id=10.1257/pol.20160589&&from=f.

C

Carrillo, Paul, Dina Pomeranz, and Monica Singhal. 2017. “Dodging the Taxman: Firm Misreporting and Limits to Tax Enforcement.” American Economic Journal: Applied Econom-ics 9 (2): 144–64.Castro, Lucio, and Carlos Scartascini. 2013. “Tax Compliance and En-forcement in the Pampas: Evidence from a Field Experiment.” IDB Working Paper IDB-WP-472 (December), Inter-American Devel-opment Bank, Washington, DC.

Cojocaru, Alexandru. 2017. “Kosovo Jobs Diagnostic.” Job Series 5, World Bank, Wash-ington, DC.

D

Del Carpio, Lucia. 2013. “Are the Neighbors Cheating? Evidence from a Social Norm Ex-periment on Property Taxes in Peru.” Work-ing paper, Princeton University, Princeton, NJ.

Dwenger, Nadja, Henrik Kleven, Imran Ra-sul, and Johannes Rincke. 2014. “Extrinsic vs Intrinsic Motivations for Tax Compliance: Ev-idence from a Randomized Field Experiment in Germany.” Paper presented at the German Economic Association’s Annual Conference 2014, “Evidence-Based Economic Policy,” Hamburg, September 7–10.

E

EBRD (European Bank for Reconstruction and Development). 2016 “Life in Transition: A Decade of Measuring Transition.” EBRD, London. https://www.ebrd.com/publications/life-in-transition-iii.

H

Hallsworth, Michael, John A. List, Robert D. Metcalfe, and Ivo Vlaev. 2017. “The Behav-ioralist as Tax Collector: Using Natural Field Experiments to Enhance Tax Compliance.” Journal of Public Economics 148 (April): 14–31.

Hernandez, Marco, Julian Jamison, Ewa Korczyc, Nina Mazar, and Roberto Sormani. 2017. Applying Behavioral Insights to Improve Tax Collection: Experimental Evidence from Po-land. Washington, DC: World Bank.

I

IMF (International Monetary Fund). 2013. Republic of Kosovo: 2013 Article IV Consulta-tion. IMF Country Report 223. Washington, DC: IMF.

K

Kettle, Stewart, Marco Hernandez, Simon Ruda, and Michael Sanders. 2016. “Behavior-al Interventions in Tax Compliance: Evidence from Guatemala.” Policy Research Working Paper 7690, World Bank, Washington, DC.

Kettle, Stewart, Marco Hernandez, Michael Sanders, Oliver Hauser, and Simon Ruda. 2017. “Failure to CAPTCHA Attention: Null Results from an Honesty Priming Experi-ment in Guatemala.” Behavioral Sciences 7 (2): 28.Kleven, H Henrik Jacobsen, Martin B. Knudsen, Claus Thustrup Kreiner, Søren Pedersen, and Emmanuel Saez. 2011. “Un-willing or Unable to Cheat? Evidence from a Tax Audit Experiment in Denmark.” Econo-metrica 79 (3): 651–92.

44


M

Mascagni, Giulia. 2015. “Tax Experiments in Developing Countries: A Critical Review and Reflections on Feasibility.” CDI Practice Paper 11 (March), Center for Development Impact, Institute of Development Studies, Brighton, U.K.

Mascagni, Giulia, Christopher Nell, and Nara Monkam. 2017. “One Size Does Not Fit All: A Field Experiment on the Drivers of Tax Com-pliance and Delivery Methods in Rwanda.” ICTD Working Paper 58 (January), Interna-tional Center for Tax and Development, Insti-tute of Development Studies, Brighton, U.K.

Ministry of Finance. 2018. “Annual Bulletin 2017: On Public Debt.” April, Debt Manage-ment Division, Ministry of Finance, Pristina, Kosovo. https://mf.rks-gov.net/desk/inc/me-dia/233CE729-B312-4F03-9081-E6193AE-2A2E9.pdf.

O

Ortega, Daniel, and Pablo Sanguinetti. 2013. “Deterrence and Reciprocity Effects on Tax Compliance: Experimental Evidence from Venezuela.” CAF Working Paper 2013/08 (December), CAF Development Bank of Latin America, Caracas.

P

Pomeranz, Dina. 2015. “No Taxation without Information: Deterrence and Self-Enforce-ment in the Value Added Tax.” American Eco-nomic Review 105 (8): 2539–69.

Pomeranz, Dina, and José Vila-Belda. 2019. “Taking State-Capacity Research to the Field: Insights from Collaborations with Tax Authorities.” Annual Review of Economics 11 (August).

R

Riinvest. 2013. To Pay or Not to Pay: A Busi-ness Perspective of Informality in Kosovo. Pris-tina, Kosovo: Riinvest Institute.

T

Torgler, Benno. 2007. Tax Compliance and Tax Morale: A Theoretical and Empirical Analysis. Cheltenham, UK: Edward Elgar Publishing.

U

UBO Consulting. 2018. “Taxpayer Survey 2017 Kosovo.” March, Gesellschaft für Inter-nationale Zusammenarbeit, Pristina, Kosovo.

W

Wenzel, Michael. 2005. “Misperceptions of Social Norms about Tax Compliance: From Theory to Intervention.” Journal of Econom-ic Psychology 26 (6): 862–83.

World Bank. 2014. “Republic of Kosovo, Public Sector Revenues: Tax Policies, Tax Evasion, and Tax Gaps.” Report 89195–XK (June), Poverty Reduction and Econom-ic Management Unit, Europe and Central Asia Region, World Bank, Washington, DC. https://openknowledge.worldbank.org/handle/10986/22269.

—

World Bank. 2015. World Development Report 2015: Mind, Society, and Behavior. Washington, DC: World Bank.

World Bank. 2017. Republic of Kosovo: Systematic Country Diagnostic. Report 114618–XK (January). Washington, DC: Europe and Central Asia Region, World Bank. http://documents.worldbank.org/curated/en/282091494340650708/pdf/Kosovo-SCD-FINAL-May-5-C-05052017.pdf.

45


Notes

1 Changes in the level and composition of taxes and government spending can affect macroeconomic variables, including economic activity, consumption, savings and investment, and income distribution. On the importance of tax collection in economic development, see “Helping Governments and Stakeholders Make the SDGs a Reality,” Sustainable Development Goals Knowledge Platform, United Nations, New York, https://sustainabledevelopment.un.org.

2 For a review of the theoretical and empirical literature on tax compliance, see Arcos Holzinger and Biddle (2016); Pomeranz and Vila-Belda (2019).

3 See Ariel (2012); Brockmeyer et al. (forthcoming); Castro and Scartascini (2013); Del Carpio (2013); Dwenger et al. (2014); Hallsworth et al. (2017); Hernandez et al. (2017); Kettle et al. (2016); Kleven et al. (2011); Ortega and Sanguinetti (2013); Pomeranz (2013); Torgler (2007); Wenzel (2005).

4 A reduced VAT rate of 8 percent is applied to the supply and import of basic foods and utilities, among other items.

5 See Ministry of Finance (2018); WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi.

6 The tax administration is considering obliging all taxpayers to submit declarations even if these exemptions are available because exemption status typically varies from one year to the next.

7 Experimental trials refers to interventions in which a comparison in outcomes across two or more groups in the context of a randomly assigned condition (in this case, receiving a reminder) could guarantee that any differences in the outcome are caused by the assigned condition rather than other factors.

8 The business-as-usual scenario in practice involves some nonrandom communication with outlier taxpayers (those with large debts, multiple missing declarations, and so on). However, during the trials, TAK suspended all communication with taxpayers within the population of interest during the design and evaluation of the trials.

9 The exclusion of taxpayers from a given trial was generally nonrandom and represented a small share of the total, typically revolving around those lacking contact information. Exclusion criteria was unique to each trial, but, in all cases, taxpayers without the relevant contact information (address, e-mail address, phone number) were excluded from the trials.

10 Randomization among the population of interest was repeated multiple times (rerandomization) to ensure a balance across treatment arms on observed characteristics during randomization. The level of randomization and use of stratification depended on the method and the heterogeneous quality of communication: clustering was necessary for e-mails and SMS messages because contact information for each method was not unique to single taxpayers. The most common explanation for this was that an e-mail or phone number presumably belonged to an accountant rather than a firm owner or individual taxpayer. Because the quality of addresses varies across the country, stratification during the PIT trial was performed at the district level to assure within-district balance. The optimal number of treatment arms was identified through power calculations using a combination of historical data and results from similar trials, though some qualitative data informed these decisions as well.

11 In the reminders, the translation of “honestly” was not deemed appropriate for the messages given the limitations of the word. “Correctly” was used in its place.

12 Letters were sent in English only. E-mails and SMS messages were sent in either Albanian or Serbian, depending on the language preference of the firm recorded in the registry. If this was not defined, the messages were sent in Albanian.

13 The small sample size meant, in most cases, that the teams were limited in their ability to test many different messages in each trial, including, for example, a reminder using standard language that had been applied in past communications.

14 When asked about the fairness of the tax system given the level of public services, 60 percent believe they should pay less in taxes; 35 percent believe they should pay the same amount of taxes; and only 5 percent think they should pay more taxes.

46


15 The Life in Transition Survey highlights the low institutional trust and the lack of satisfaction with social services (EBRD 2016). Notwithstanding this mistrust, a large share of the population, over 80 percent, are willing to give more money to improve social services and help the needy, and attitudes toward government spending priorities mostly revolve around problems in social services and the assistance to the poor.

16 The population of interest also included taxpayers who were still in the system despite their inactivity since 2010. These individuals were identified through their unique fiscal numbers.

17 The random assignment to one of the three groups was carried out using the e-mail addresses rather than the taxpayer names because each message that would make reference to all corresponding unique fiscal numbers would be sent to only one e-mail address.

18 The random assignment to one of the three groups was carried out at the level of cell phone number rather than the taxpayer name because each SMS message could only be communicated to a cell phone.

19 In Albanian, the simple reminder with the social norms message read as follows: “Mesazh nga ATK: deklaroni, paguani TVSH Gusht 2018 me korrektësi, me kohë! 7 nga 10 firma i dorëzojnë deklaratat e tyre me kohë. Mos prisni, jeni pjesë e SHUMICËS!”

20 Statistically significant here and throughout the text indicates that the causality link is at least 95 percent likely to be true. Statistically insignificant means that causality cannot be confirmed with this degree of certainty, though it does not imply that the there is no causality.

21 Robustness checks were carried out that included adding various controls to the estimate of treatment impacts and excluding certain groups from the regressions.

22 The impact of the letters on the declaration rate by April 13th, the due date highlighted in the letters, was 2 percentage points for the simple + deliberate choice letter and 3 percentage points for the simple letter (both statistically higher than the group that received no reminder).

23 Because treatment noncompliance was one-sided (some taxpayers assigned to treatment did not receive the letter, but no taxpayers assigned to the control group received a letter), the local average treatment effect is reduced to the average treatment on the treated. Thus, because many of those who should have received a letter did not receive one, the total effect of the treatment is the effect on those who did receive the letter.

24 VAT declarations for a given month are filed the following month; so, the declaration for June was to be filed in July.

25 Moreover, the declaration process (which has entirely migrated to e-filing) was highlighted as being generally easy to manage for the average taxpayer. Forms are simple and use relatively clear and concise language. As a result, the experiments did not consider form-simplification to reduce the cognitive burden of submitting tax declarations.

47


Annex

Figure A.1. MACROECONOMIC INDICATORS FOR KOSOVO, 2012–18

Income and economic growth 2012 2013 2014 2015 2016e 2017e 2018f

GDP growth (annual %) 2.90 3.43 1.20 4.09 4.06 4.23 4.46

Composition

Consumption 6.78 3.84 5.45 2.70 4.70 1.49 1.18

Private consumption 6.41 3.30 5.86 3.76 5.67 1.57 0.85

Public consumption 0.37 0.54 −0.41 −1.06 −0.97 −0.08 0.33

Investments −4.37 −0.10 −1.37 2.90 2.07 1.52 3.52

Net exports 0.48 −0.32 −2.88 −1.51 −2.71 1.21 −1.85

Exports −3.92 −0.38 1.39 0.43 0.52 3.97 3.11

Imports 4.40 −6.52 −4.27 −1.94 −3.23 −2.76 −4.96

GDP per capita growth (annual %) 2.03 2.36 1.33 5.25 3.23 3.40 3.63

GDP per capita (current US$ millions) 3,600.87 3,877.96 4,054.64 3,745.75 3,698.47 3,896.96 4,046.76

Private consumption growth (annual %) 7.75 3.95 6.98 4.34 6.63 1.83 2.90

Gross investment (% of GDP) 26.03 24.82 23.24 25.81 25.54 27.23 29.64

Gross investment - public (% of GDP) 10.88 9.93 7.39 6.95 7.31 7.41 9.13

Gross investment - private (% of GDP) 15.33 14.78 15.69 17.98 17.33 19.83 20.51

Note: e = estimate. f= forecast.

1) Fiscal balance includes the regular fiscal deficit within the 2 percent fiscal rule (lower than 2 percent projections here) and the deficit due to investment clause

Sources: Ministry of Finance, Central Bank of Republic of Kosovo, Agency of Statistics of Kosovo, World Bank calculations.

48


Figure A.2. FISCAL INDICATORS FOR KOSOVO, 2012–18

2012 2013 2014 2015 2016e 2017e 2018f

Fiscal indicators

Revenue (% of GDP) 26.13 24.64 23.95 25.08 26.30 26.83 26.98

Expenditure (% of GDP) 28.49 27.94 26.60 27.08 27.75 28.08 30.16

Current (% of GDP) 17.37 17.75 18.95 19.85 20.13 20.34 20.91

Capital (% of GDP) 11.09 9.82 7.39 6.95 7.19 7.44 8.99

Overall fiscal balance (% of GDP) /1

−2.37 −3.30 −2.65 −1.99 −1.45 −1.24 −3.18

Primary fiscal balance (% of GDP) /1

−2.12 −3.04 −2.38 −1.71 −1.13 −0.95 −2.84

Overall fiscal balance per regular budget (% of GDP) /1

−2.37 −3.30 −2.65 −1.99 −1.45 −1.21 −1.76

Total public and publicly guaranteed debt (% of GDP)

8.10 9.01 10.56 12.83 14.41 16.51 17.66

Total public debt (% of GDP)

8.10 9.01 10.39 12.66 14.08 15.81 16.91

Domestic public debt (% of GDP)

1.45 2.86 4.61 6.50 7.89 9.09 10.36

External public debt (% of GDP)

6.65 6.15 5.78 6.16 6.19 6.72 6.54

Note: e = estimate. f= forecast.

1) Fiscal balance includes the regular fiscal deficit within the 2 percent fiscal rule (lower than 2 percent projections here) and the deficit due to investment clause.

Sources: Ministry of Finance, Central Bank of Republic of Kosovo, Agency of Statistics of Kosovo, World Bank calculations.

49


Figure A.3a. SIMPLE REMINDER LETTER (ALBANIAN)

50


Figure A.3b. SIMPLE REMINDER LETTER (ENGLISH TRANSLATION)

51


Figure A.4a. SIMPLE REMINDER WITH DELIBERATE CHOICE MESSAGE LETTER (ALBANIAN)

52


Figure A.4b. SIMPLE REMINDER WITH DELIBERATE CHOICE MESSAGE LETTER (ENGLISH TRANSLATION)

53


FIGURE A.6A. SIMPLE REMINDER WITH POSITIVE FRAMING E-MAIL (ALBANIAN)

54


Figure A.6b. SIMPLE REMINDER WITH POSITIVE FRAMING E-MAIL (ENGLISH TRANSLATION)

55


Figure A.5a. SIMPLE REMINDER E-MAIL (ALBANIAN)

56


Figure A.5b. SIMPLE REMINDER E-MAIL (ENGLISH TRANSLATION)

World Bank Publications, The World Bank Group. PROMOTING TAX COMPLIANCE IN KOSOVO WITH BEHAVIORAL INSIGHTS

Washington, DC. March 2019.

Date post:	06-Nov-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

PROMOTING TAX COMPLIANCE IN KOSOVO WITH BEHAVIORAL … · 2019. 3. 5. · 03 Kosovo Promoting Tax...

Documents